v3.3.8

What's Changed

• [FEAT] Add F1 score to ROC and precision/recall charts by @NickCrews in #807

• [MAINT] Add test from discussion 799 by @RobinL in #803

• [FIX] Add preceding blocking rules to eliminate dupes in find_matches_to_new_records by @RobinL in #810

New Contributors

• @NickCrews made their first contribution in #807

Full Changelog: v3.3.7...v3.3.8

v3.3.7

What's Changed

• [FIX] Improve poor performance of linker.prediction_errors_from_labels_table in Spark by @RobinL in #801

Full Changelog: v3.3.6...v3.3.7

v3.3.6

What's Changed

• [FIX] fix predict and match weight parts by @RobinL in #797

Full Changelog: v3.3.5...v3.3.6

v3.3.5

What's Changed

• [fix] 795 too many tf columns selected by splink by @RobinL in #796

Full Changelog: v3.3.4...v3.3.5

v3.3.4

What's Changed

• [FIX] Fix errors from missing columns when calling ROC/Precision recall chart from labels table by @RobinL in #794

Full Changelog: v3.3.3...v3.3.4

v3.3.3

What's Changed

• [DOCS] Add febrl3 example to docs by @RobinL in #790
• [FIX] Fix problem with identifying _is_exact_match rules by @RobinL in #791

Full Changelog: v3.3.2...v3.3.3

v3.3.2

What's Changed

• [MAINT] Remove print statement by @RobinL in #789

Full Changelog: v3.3.1...v3.3.2

v3.3.1

What's Changed

• [FIX] Ensure pipeline resets by @RobinL in #787

• [DOCS] Remove realtime from tutorial by @RobinL in #782

• [DOCS] 50k example by @RobinL in #783

• [DOCS] Add examples index by @RobinL in #784

• [MAINT] Bump version to v3.3.1 by @RobinL in #788

Full Changelog: v3.3.0...v3.3.1

v3.3.0

What's Changed

Features

• [FEAT] Add percentage difference to comparison level library by @RobinL in #757
• [FEAT] Add jaro winkler to duckdb linker now 0.5.0 is a dependency by @RobinL in #766
• [FEAT] Waterfall of false positives and false negatives from labels by @RobinL in #763

Other

• [DOCS] Add links to videos into main readme by @RobinL in #765
• [DOCS] Add examples section to docs by @RobinL in #772
• [FEAT] ROC/Precision recall/truth table from label column name by @RobinL in #773
• [DOCS] Add examples to docs by @RobinL in #774
• [FIX] Fix jaro in duckdb by @RobinL in #775
• [FIX] Jaro spark fix by @RobinL in #776
• [DOCS] Add QA from ground truth (cluster) column to docs by @RobinL in #777
• [MAINT] Bump version to 3.3.0 by @RobinL in #778

Note there's a small backwards incompatible change to the API where previous earlier names for accuracy functions have been replaced with the following:

The underlying data table for ROC/precision recall analysis
linker.truth_space_table_from_labels_column()
linker.truth_space_table_from_labels_table()

ROC charts
linker.roc_chart_from_labels_column()
linker.roc_chart_from_labels_table()

Precision recall charts
linker.precision_recall_chart_from_labels_column()
linker.precision_recall_chart_from_labels_table()

Individual predictions which are either false positives or false negatives, to plot in waterfall chart
linker.prediction_errors_from_labels_column()
linker.prediction_errors_from_labels_table()

Full Changelog: v3.2.1...v3.3.0

v3.3.0.dev03

What's Changed

• [FIX] Jaro spark fix by @RobinL in #776

Full Changelog: v3.3.0.dev02...v3.3.0.dev03